Journal of Cardiovascular Magnetic Resonance (2006) 8, 335–344Copyright c© 2006 Taylor & Francis Group, LLCISSN: 1097-6647 print / 1532-429X onlineDOI: 10.1080/10976640500451945

MYOCARDIAL INFARCTION

Utility of Contrast-Enhanced Cardiovascular MagneticResonance (CE-CMR) to Assess How Likely is an

Infarct to Produce a Typical ECG PatternJuan Manuel Cino,1 Sandra Pujadas,2 Francesc Carreras,2 Iwona Cygankiewicz,1 Ruben Leta,2 Mariana Noguero,2

Xavier Garcia-Moll,2 Tony Bayes Genıs,2 Guillem Pons-Llado,2 and Antoni Bayes de Luna1

Institut Catala de Ciencies Cardiovasculars, Hospital Sant Pau, Sant Antoni M, Claret 167, CP 08025, Barcelona, Spain,1

Servei de Cardilogia, Hospital de Sant Pau, Barcelona, Spain2

ABSTRACT

Objectives: For over 50 years, Q-wave myocardial infarction (MI) location has been basedon pathologic ECG studies. Although contrast-enhanced magnetic resonance (CE-CMR) is cur-rently the “gold standard” technique for location and quantification of necrotic areas, we foundno large study in the literature devoted to establish which ECG patterns corresponds to differ-ent MI location detected by CE-CMR. We hypothesized that CE-CMR would be very accurate forevaluating different ECG patterns and its sensitivity (SE) and specificity (SP) for locating MI indifferent LV areas. Methods and results: CE-CMR/ECG correlation was studied in 48 patientswho presented a first MI due to acute coronary syndrome (ACS) with ST-segment elevation andin whom CE-CMR was performed in chronic phase. We evaluated the ECG patterns that bestcorrelated with the 7 prespecified necrotic areas assessed by CE-CMR, 4 in anteroseptal zone(septal, apical/anteroseptal, extensive anterior, and limited anterolateral) and 3 in inferolateralzone (inferior, lateral and inferolateral). The global concordance between CE-CRM and ECGwas of 75% and 7 ECG patterns were stablished. Conclusion: The capacity of CE-CMR to detectECG patterns for necrotic area location presents highly acceptable concordance. Thanks toCE-CMR, we defined 7 ECG patterns for MI detection according to the 7 areas of the LV studied.The areas that present more cases with normal ECG are limited anterolateral and the areas ofthe inferolateral zone.

INTRODUCTION

Q-wave myocardial infarction (MI) location based on patho-logic ECG studies conducted by Myers in the 1940s (1–2) be-came with minor changes, the most popular classification of Qwave MI (3–9). According to this classification, the presenceof Q-waves in V1-V2 corresponds to septal MI, in V3-V4 toanterior MI, in V5-V6 and/or I-VL to low and high lateral MI,respectively, in II, III and VF to inferior or diaphragmatic MI,and R-(RS) in V1–V2 a mirror pattern of posterior infarction.

Keywords: Contrast-Enhanced Magnetic Resonance,Electrocardiography, Q Wave InfarctionCorrespondence to:A. Bayes de LunaCatedratico de Cardiologıa de la UABInstitut Catala de Ciencies Cardiovasculars (ICCC)Sant Antoni M Claret 167, CP 08025, Barcelona, Spainemail: abayesluna@santpau.es

However, clear evidence exists of the limitations of this strictclassification owing to difficulties in correlating precordial leadswith affected walls (precordial electrodes are often not wellplaced) and to changes in the precordial lead-heart wall relation-ship in subjects with different body builds (10–12). Furthermore,a large amount of information has been obtained from elec-trophysiological (13), hemodynamic, and angiographic studies(14–16) and, particularly, correlations with imaging techniques(17–19). Contrast-enhanced cardiovascular magnetic resonance(CE-CMR) has become the “gold standard” imaging techniquefor identifying the area of necrosis (20–24). Therefore, it is theideal technique for asessing which ECG patterns better matchthe different infarcted areas detected by CE-CMR. The resultsof this study are presented herein.

MATERIAL AND METHODS

CE-CMR and ECG findings were analyzed in 48 patients(40 men, 8 women, mean age: 62 yrs) in the chronic phase of

335

MI. All had presented an ST elevation ACS. All patients werereperfused, with fibrinolytic or interventional therapy. Coronaryangiography or multislice scanning was performed in 90% ofcases.

ECG recordings

Standard 12-lead ECGs were recorded at 25 mm/s speed and10 mm/1 mV voltage in the chronic phase of MI (≥ 3 weekspost MI). The ECG recordings were reviewed by 2 independentinvestigators blinded to clinical and CMR data. In cases of dis-crepancy, the final decision was made by a third investigator. MIwas diagnosed according to the following ECG criteria (3, 4, 9,25, 26) :1) any Q-wave ≥30 ms in inferior/lateral leads; 2) Qwave ≥40 ms in I, VL; 3) Q-wave in ≥2 contiguous precordialleads; or 4 any Q-wave in V1-V2 or R wave ≤0.1 mV in V2.The presence of RS in V1, R in V1 ≥40 ms, and “qr” or lowvoltage “r” <5 mm, V6 were considered Q wave equivalents.

Cardiovascular magnetic resonance studies

A CE-CMR study was performed with a Philips Intera 1.5Tscanner (Best, The Netherlands) in all patients. After the usualscout planes had been obtained, steady-state free-precessioncine-MR images were acquired in individual long-axis planesand in multiple 10 mm thick short axis slices from the atrioven-

tricular ring to the apex of the left ventricle. Sixteen phases ofthe cardiac cycle were acquired for each slice and displayed asa loop.

Intravenous gadobutrol (Gadovist,©R Schering AG, Berlin,

Germany) was injected at a dose of 0.1 mmol/kg. A 3D inversion-recovery segmented gradient echo sequence was acquired 10minutes after contrast administration to assess delayed contrastmyocardial hyperenhancement (HE). Inversion times were ad-justed to null the signal from normal myocardium (200–300 ms).This sequence was prescribed in multiple short axis planes usingthe same orientation as the cine-MR images and acquired duringa patient breath-hold of approximately 20 seconds.

MI was considered transmural if at least 1 segment of theNorth America Society of Imaging (NASI) statement (27)fulfilled the CMR criteria for being transmural (HE in > 50%of wall thickness). To correlate CMR and ECG findings, wedivided the left ventricle into two zones (Fig. 1): anteroseptalwith some lateral and usually also inferior involvement,perfused by the left anterior descending artery (LAD) and itsbranches, and inferolateral perfused by the right coronary artery(RCA) and left circumflex artery (LCX) and its branches. Eachzone was then divided into different areas and segments (Fig. 1)according to the NASI statement (27). Ce-CMR areas wereidentified by 3 experts in imaging, and its correspondence withthe myocardial NASI segments was determined. According tothe presumed site of coronary occlusion, four areas of infarction

Figure 1. A: Bull’s eye image of the left ventricle with its division into two main zones: anteroseptal (irrigated by LAD) and inferolateral (irrigatedby RCA or LCX). The area with mixed irrigation is represented in grey. B to D: Bull’s eye image of the left ventricle and the perfusion of its wallsand segments according to North American Societies of Imaging statement.

336 J. M. Cino et al.

were defined in the anteroseptal zone affecting: 1) septum andpart of the anterior wall (A1); 2) apical area with or withoutextension to upper but not basal part of the anterior and septalwalls (A2); 3) extensive involvement of the anterior and septalwall also with lateral involvement (A3); and 4) limited partof the anterior wall, especially the middle segment often withsome extension to the middle/low lateral wall (A4). Three areasof infarction were defined in the inferolateral zone affecting:1) lateral wall (B1); 2) inferior wall (B2)∗ and 3) inferolateralwall (B3). The bull’s eye distribution of these areas with all thesegments involved according to the statement of NASI (27) areshown in Table 1. The seven identified areas were correlatedwith ECG recordings with the aim of establishing seven ECGpatterns that better matched each prespecified area of infarction.

Statistics

Q-wave MI and CE-CMR location were described by meansof percentage. Categorical variables were presented as propor-tions. To define the ECG patterns that better matched with theseven predefined infarction areas detected by CE-CMR, we cal-culated the sensitivities and specificities of different ECG pat-terns by using 2 × 2 contingence tables. In each of these tables,the presence or absence of CE-CMR location was correlated withthe presence or absence of Q-waves or equivalents in differentleads. Finally, we defined seven ECG patterns that presented thehighest sensitivity and specificity and global concordance thatwas calculated by using the Kappa index.

RESULTS

Q/non-Q-wave MI

Forty of 48 patients presented Q-wave MI or equivalents inchronic phase. The ECG in the other 8 cases could be considerednormal or borderline (2 with positive T wave in V1 and 1 withslurrings at the end of QRS) (see Table 1).

Correspondence between CMR/ECGfor MI location

MI were grouped according to the 7 predefined areas of in-farction as shown by CE-CMR and 7 ECG patterns that bettermatched with these areas. Table 1 shows this relationship withthe corresponding ECG patterns sensitivity and specificity andthe name given to the infarction. The global concordance be-tween CE-CMR and ECG was 75%. Some examples of CMRimages and their corresponding ECG patterns are displayed inFigs. 2–5. Three cases with limited anterolateral infarction and 5

∗Inferior wall corresponds to segments 4,10 and 15 according tothe statement of NASI (Cerqueira 2002). Usually, ECG books refer toinferoposterior wall, with segment 4 being named posterior and in theNASI statement inferobasal (see Discussion). In this article, the terminferior wall will be used instead of inferoposterior, and inferobasalsegment instead of posterior segment (n◦ 4).

cases of MI of the inferolateral wall infarctions presented normalECG in the chronic phase.

DISCUSSION

In this study, we present 7 ECG patterns that may be usefulto locate myocardial necrosis as assesed by CE-CMR.

Q-wave MI location based on correlation of pathologic ECGstudies has several limitations owing to technical aspects, and,as a consequence, these correlations are being performed only infew cases, usually in patients with extensive MI. The correspon-dence between ECG findings and recent imaging techniqueshas become crucial to overcome these limitations. Echocardio-graphy is the easiest comparative diagnostic imaging technique(17, 18, 28, 29). However, echocardiography frequently overesti-mates the infarcted area; therefore, its reliability is not excellent.Scintigraphy is very reliable to detect ischemic (18) and infarctedareas (30), but it has been demonstrated that the degree or eventhe presence of some types of necrosis (non-Q-wave infarction)may be underestimated (31). Recently, CE-CMR after gadolin-ium injection has permitted an accurate identification of necroticareas when compared to pathologic studies (32). Furthermore,in clinical trials CE-CMR has been shown to be extremely reli-able in assessing not only the extension of the infarcted area butalso its transmurality (20–21, 24). Thus, CE-CMR has becomethe “gold standard” method for quantifying necrotic myocardialmass and also for differentiating between transmural and non-transmural infarcted areas in the chronic phase of ischemic heartdisease (20).

To our knowledge, the correlation between CE-CMR in-farcted area and Q-wave has only been performed partially andwith a small number of patients (23, 24). These studies focusmainly on detection and differentiation between transmural andnon-transmural myocardial infarction, and its correspondenceto the appearence of Q-wave in the surface ECG considered asan equivalent of transmural compromise. No attention was paidto the correlation between exact localization of MI based onthe leads where Q-waves were observed and CE-CMR findings.However, such a correlation is useful not only for academicsbut also for everyday clinical practice. Ascertaining the areainvolved through the ECG pattern has important clinical im-plications and adds more information to quantification of theinfarcted myocardial mass.

We agree with the previously presented consensus of theNorth American Imaging Techniques Societies (27) that dividedthe heart into 17 segments corresponding to the four heart walls(anterior, septal, lateral, and inferior). However, we believe thatthe decision of this consensus to give the name inferior wallto what is really inferoposterior wall must be discussed. Seg-ment 4 is named differently by electrocardiologists (posterior)or imaging experts (inferobasal). As may be demonstrated byCE-CMR studies, the basal part of this wall sometimes archesmore upwards and therefore converts into an authentic poste-rior position. Furthermore, this posterior part may be smaller orlarger according to the position of the heart in the thorax. In veryvertical hearts, it may be almost exclusively posterior. According

CE-CRM-ECG Correlation for Location of Q Wave Infarction 337

Table 1. Correlations between the different myocardial infarction (MI) types with its infarction area assesed by CMR, ECG pattern and name givento the infracrtion. The grey zone and arrows seen in bull’s eye correspond to infarction areas and its possible extension. Two cases of A-4; three ofB-1; one of B-2 and two of B-3 presented normal ECG

to terminology used in electrocardiography for over 60 years,the term posterior infarction, and consequently also posterior is-chemia and injury, is currently used to locate the infarcted areain the basal part of the inferoposterior wall (mainly segment 4),usually due to occlusion of the non-dominant LCx artery. Nev-ertheless, it is true that infarction of the basal part (segment 4)of the inferoposterior wall (inferior according Cerqueira’s state-

ment; 27) does not present RS in V1 and really this morphologyis found in infarction of the inferior part of the lateral wall (seg-ments 5 and 11). Indeed, our results confirm the findings of otherauthors (21) who already found that lateral infarction originatespositive R waves in V1. Thus, R (RS) in V1 does not corre-sponds to segment 4 because it is really due to lateral infarction(necrosis of segments 5 and 11) and not to necrosis of segment 4.

338 J. M. Cino et al.

Figure 2. ECG and CE-CMR images obtained in a patient with apical MI (A2 type). A) Horizontal long-axis view: myocardial HE (bright area)indicates mid-low and apical septal necrosis. B) Vertical long-axis view: HE shows apical necrosis involving the anterior and the inferior wall atthis level. C) Short-axis view at the mid level: the absence of myocardial HE at this level indicates that the middle septum is spard. D) Short-axisview at the apical level: HE corresponding to an inferior and septal myocardial necrosis. Notice that at this level part of the anterior wall alsoshows HE indicating necrosis.

CE-CRM-ECG Correlation for Location of Q Wave Infarction 339

Figure 3. ECG and CE-CMR in limited anterolateral MI (A4 type). QR in VL and low voltage “R” in V2 is shown. A) Horizontal long-axis view:myocardial HE (arrows) indicates non-transmural necrosis of the lateral wall at the mid level; B) Vertical long axis view: Myocardial HE (arrows)shows necrosis of the mid anterior wall. At the bottom, short axis images shows myocardial HE (arrows) at the two levels of the mid-anterior wall(C, D), whereas no HE is observed at apical (E) levels.

We consider that segment 4 may be named as suggested byCerqueira (27) inferobasal instead of posterior, and that positiveR wave in V1 is not a criterion of posterior MI but of lateral MI.

It has been considered that the ECG abnormalities observedin II, III, VF leads (ST elevation, negative T wave or Q wave

of necrosis) without RS in V1 correspond to involvement of theinferior wall. However, frequently there is involvement not onlyof inferior wall (segments 4,10,15) but also of the inferior part ofthe septum (segments 3, 9, and 14). In these cases, the occludedartery is usually the RCA. When the ECG presents Q in II, III

340 J. M. Cino et al.

Figure 4. ECG and CE-CMR in lateral (B1) MI. QR in VL with low voltage “r” in I and V6 and RS in V1 (with wide-slurred R) is the characteristicECG pattern seen in some cases of lateral necrosis. Horizontal long-axis view (A): myocardial HE (arrows) shows non-transmural necrosis ofthe lateral wall at the basel and mid level. Short axis images at the basal (B) and mid level (C): myocardial HE (arrows) indicates non-transmuralnecrosis of the lateral wall. At the apical level (D) no myocardial HE is observed.

and VF and RS in V1, the involved area is the inferior wall andinferior part of the lateral wall with or without involvement ofthe inferior part of the septum.

The limited anterolateral MI (A-4 type) must be differentiatedfrom the lateral MI (B-1) owing to the frequent presence in bothof Q wave in I and VL and from septal MI (A-1) owing to thepossible presence of q in V2-V3 in both. The presence of lowQR not QS in VL often with QR pattern or low R in V5-6 favorslateral MI (B-1), and the presence of QS or QrS in V1 and thelack of pathologic Q wave in VL favors septal MI (A-1).

It has been demonstrated thanks to CE-CMR (22, 24),echocardiographic, angiographic, or scintigraphic (14–17) cor-relations, that QS morphology observed in VL (and sometimesin I) is due to infarction produced by LAD subocclusion involv-ing diagonal arteries. In these cases, the necrosis is located inthe mid-anterior wall often with certain propagation to the mid-low anterior part of the lateral wall. This may be explained bythe fact that this area is perfused by the first diagonal artery,while the inferior part and the majority of the basal-anterior part

of the lateral wall is perfused by the LCX (OM) or intermedi-ate artery. Therefore, the term “high lateral infarction” appliedto QS in VL (and sometimes in I lead) is confusing since theabove-mentioned morphology appears in the absence of necro-sis of the basal area of this wall. On the other hand, when theinfarction involves mainly the lateral wall (occlusion of the ob-tuse marginal artery), “QR” is more frequently seen than “QS”morphology and is also seen in V5-V6.

According to the classical concept, the QS pattern recordedfrom V1 to V4 was considered as anteroseptal infarction, whilecontemporary knowledge demonstrates that it represents apicalinvolvement (19). However, it is clear that many cases of isolatedapical infarction do not present QS in V1 to V4 (33). Therefore,this criterion is highly specific for apical involvement but rela-tively insensitive. In our study we included cases of apical infarc-tion with or without anterospetal extension in one group (A-2,named apical/anteroseptal) as in these cases we have not foundany differences in the ECG pattern in precordial leads. This ispartially due to the fact that the first vector of depolarization,

CE-CRM-ECG Correlation for Location of Q Wave Infarction 341

Figure 5. ECG and CE-CMR in inferior MI (B2 type). The ECG shows typical qR pattern in III, and VF with rS in V1 and RS in V2 with tall peakedT. Note that despite clear inferobasal necrosis (gadolinium HE in A), V1 shows rS pattern ant not RS (as seen in some cases of lateral infarction).Images of the short-axis view shows myocardial HE (arrows) at the basal (A) and mid (B) levels corresponding to transmural necrosis of theinferior wall at the basel and mid level. No myocardial HE is observed at the apical level indicating absence of myocardial necrosis(C).

responsible for R in V1-V3 is originated in the mid-low part ofseptum (13). Probably, in case of isolated apical involvement, itis more frequent to find Q-waves in inferior leads. Therefore, inour experience QS from V1-2 to V5-6 may be found in apicalinfarction with and without anteroseptal extension.

We found a good correspondence between the infarcted areadetected by CE-CMR and ECG pattern (75% of global con-cordance), considering that cases with normal ECG were also

included. In fact, when we assessed, in other study in process,the accuracy of MI location by the leads containing Q waves ofequivalent (from ECG to CE-CMR), we found that the correla-tion is even better.

Nowadays, due to widely used reperfusion techniques, espe-cially in case of thrombolytic treatment, it is impossible to besure about the place of occlusion in the culprit artery in the acutephase of myocardial infarction while assessing the angiographic

342 J. M. Cino et al.

findings in the chronic phase of MI. Nevertheless, it can be de-ducted basing on the already performed correlations betweenECG and angiographic findings in the acute phase (10). In thiswork we did not focus on determination of the culprit artery orthe place of its occlusion. Attention was paid to necrosis as thefinal consequence of coronary occlusion assessed by CE-CMRand its electrocardiographic expression.

LIMITATIONS

Although these results are very encouraging regarding thecorrespondence between CE-CMR technique as the gold stan-dard and ECG for MI location, we must acknowledge the limi-tations, owing to the reduced number of cases and the fact thatonly cases with ACS and ST elevation in the acute phase of MIwere included. Hence, a larger prospective study is required fora definitive statement based on these correlations to be made.

CONCLUSIONS

1. Seven pre-specified areas of myocardial infarction detectedby CMR technique had good concordance (75%) with 7 ECGpatterns (Table 1). 2. Seventeen percent of MI detected by CE-CMR presented normal ECG. 3. Infarction of the lateral wall(especially segments 5 and 11) and not of the inferobasal partof the inferior wall (segment 4) explains RS morphology in V1.4. Infarction of mid anterior and mid-lateral wall and not of highlateral wall is responsible of QS morphology in VL (I).

ACKNOWLEDGEMENT

We would like to thank to Creu Blanca Medical Center,Barcelona.

REFERENCES1. Myers G, Howard AK, Stofer BE. Correlation of electrocardio-

graphic and pathologic findings in anterior infarction. Am HeartJ 1948;36:535–50.

2. Myers G, Howard AK, Stofer BE. Correlation of electrocardio-graphic and pathologic findings in lateral infarction. Am Heart J1948;37:374–417.

3. Bayes de Luna A. Textbook of Clinical Electrocardiography. 2nded. Futura Publishing: Armonk, 1998, pgs. 121–157.

4. Chou T. Electrocardiography in clinical practice. Grune & Stratton:Nueva York, 1979.

5. Fisch CH. Electrocardiography. In: Braunwald’s textbook on heartdiseases, 5th edition. Braunwald E, ed. Mexico: McGraw-Hill Inter-americana, 1999, pgs 115–165.

6. Lipman BS, Massie E, Kleiger ZE. Clinical scale electrocardiogra-phy. Year Book Medical Publisher: Chicago, 1972, pgs 218–225.

7. McFarlane P, Veitch L. Comprehensive electrocardiology. Perga-mon Press: New York, 1989.

8. Sodi Pallares D, Bisteni A, Medrano G, Ayola C. Electrocardiogra-phy and vectorcardiography. Grune—Stratton: New York, 1960.

9. Wagner GS, Marriot’s Electrocardiography. Lippincot Williams andWilkins: Philadelphia, 2001, pgs 184–192.

10. Bayes de Luna A, Fiol M. Imagen de isquemia, lesion y necrosis.La electrocardiografia de la cardiopatia isquemica. Correlaciones

clınicas e implicaciones pronosticas. Prous Science: Barcelona-Philadelphia. In press.

11. Goldberger E. Unipolar lead ECG and VCG. Krimpton Publisher:London, 1953, pgs 284–316.

12. Massie E, Walsh T. Clinical ECG and vectocardiography. The YearBook Medical Publisher: Chicago, 1960, pgs 268–317.

13. Durrer D, Van Dam R, Freud G, Janse M, Meijler F. Total excitationof the isolated human heart. Circulation 1970;41:899–910.

14. Shen W, Tribouilloy C, Lesbre JP. Relationship between elec-trocardiographic patterns and angiographic features in isolatedleft circumflex coronary artery disease. Clin Cardiol 1991;14:720–4.

15. Takatsu F, Osugui J, Nagaya T. Is it possible to rule out extensiveanterior myocardial infarction in the absence of abnormal Q wavein lead I and aVL? Effect of inferoapical extention of infarction overapex. Jpn Circ J 1986;50:601–6.

16. Warner R, Hill NR, Mookherjee S, Smulyan H. Diagnostic sig-nificance for coronary artery disease of abnormal Q-waves inthe “lateral” electrocardiographic leads. Am J Cardiol 1986;58:431–5.

17. Bogaty P, Boyer L, Rousseau L, Arsenault M. Is anteroseptal my-ocardial infarction an appropriate term? Am J Med 2002;113:37–42.

18. Gallik DM, Obermueller SD, Swarna US. Simultaneous assess-ment of myocardial perfusion and left ventricular dysfunction duringtransient coronary occlusion. J Am Coll Cardiol 1995;25:1529–35.

19. Shalev Y, Fogelman R, Oettinger M, Caspi A. Does the electro-cardiographic pattern of anteroseptal myocardial infarction corre-late with the anatomic location of myocardial injury? Am J Cardiol1995;75:763–6.

20. Engblom H, Wagner G, Setser R, Selvester R, Billgren T, KrasperJ, Maynard C, Olle P, Arheden H. Quantitative clinical assessmentof chronic anterior myocardial infarction with delayed enhance-ment magnetic resonance imaging and QRS scoring. Am Heart J2003;146:359–66.

21. Moon JC, De Arenaza DP, Elkington AG, Taneja AK, John AS,Wang D. The pathologic basis of Q-wave and non-Q-wave my-ocardial infarction: a cardiovascular magnetic resonance study. JAm Coll Cardiol 2004;44:554–60.

22. Selvanayagam J, Kardos A, Nicolson D, Francis J, Petersen S,Robson M, Banning A, Neubauer S. Anteroseptal or apical myocar-dial infarction: a controversy addressed using delayed enhance-ment cardiovascular magnetir resonance imaging. J CardiovascMagn Reson 2004;6:653–61.

23. Sievers B, John B, Brandts B, Franken U, van Bracht M, TrappeHJ. How reliable is electrocardiography in differentiating trans-mural from non transmural MI? A study with contrast magneticresonance imaging as gold standard. Int J Cardiol 2004;97:417–23.

24. Wu E, Judd RM, Vargas JD, Klocke FJ, Bonow RO, Kim RJ.Visualization of presence, location, and transmural extent ofhealed Q-wave and non-Q-wave myocardial infarction. Lancet2001;357:21–8.

25. Antman EM, McCabe CH, Gurfinkel EP, Turpie AG, Bernink PJ,Salein D, Bayes De Luna A, Fox K, Lablanche JM, Radley D,Premmereur J, Braunwald E. Enoxaparin prevents death and car-diac ischemic events in unstable angina/non-Q-wave myocardialinfarction. Results of the thrombolysis in myocardial infarction(TIMI) 11B trial. Circulation 1999;100:1593–601.

26. Alpert JS, Thygesen K, Antman E, Bassand JP. Myocardial in-farction redefined—a consensus document of The Joint EuropeanSociety of Cardiology/American College of Cardiology Commit-tee for the redefinition of myocardial infarction. J Am Coll Cardiol2000;36:959–69.

27. Cerqueira MD, Weissman NJ, Dilsizian V. Standardized myocardialsegmentation and nomenclature for tomographic imaging of theheart: A statement for healthcare professionals from the Cardiac

CE-CRM-ECG Correlation for Location of Q Wave Infarction 343

Imaging Committee of the Council on Clinical Cardiology of theAmerican Heart Association. Circulation 2002;105:539.

28. Matetzky S, Freimark D, Feinbergt MS. Acute myocardial infarctionwith isolated ST-segment elevation in posterior chest leads V7-9.“Hidden” ST-segment elevations revealing acute posterior infarc-tion. J Am Coll Cardiol 1999;34:748–54.

29. Porter A, Sclarovsky S, Ben-Gal T. Value of T-wave direction withlead III ST-segment depression in acute anterior wall myocardialinfarction: electrocardiographic prediction of a “wrapped” left ante-rior descending artery. Clin Cardiol 1998;21:562–5.

30. Zafrir B, Zafrir N, Gal T, Adley Y, Iakobishvili Z, Rahman M,Birnbaum Y. Correlations between ST elevation and Q waves onthe predischarge electrocardiogram and the extend and location ofMIBI perfusion defects in anterior myocardial infarction. Ann Non-invasive Electrocardiol 2004;9:101–112.

31. Wagner A, Mahrholdt H, Holly T, Elliott M, Regenfus M, ParkerM, Klocke F, Bonow R, Kiim R, Judd R. Contrast enhancedCMR and routine single photon emission computed tomogra-phy (SPECT) perfusion imaging for detection of subendocar-dial myocardial infarcts: an imaging study. Lancet 2003;361:374–9.

32. Kim RJ, Fieno DS, Parrish TB, Harris K, Chen EL, SimonettiO, Bundy J, Finn JP, Klocke FJ, Judd RM. Relationship ofMRI delayed contrast enchancement to irreversible injury, in-farct age, and contractile function. Circulation 1999;100:1992–2002.

33. Giannuzzi P, Imparato A, Luigi P, Santoro F, Tavazzi L. Inaccuracyof various proposed electrocardiographic criteria in the diagno-sis of apical myocardial infarction—a critical review. Eur Heart J1989;10:880–6.

344 J. M. Cino et al.